In [6]:
%matplotlib inline
import warnings
warnings.filterwarnings("ignore")
import sqlite3
import pandas as pd
import numpy as np
import nltk
import string
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.feature_extraction.text import TfidfTransformer
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.feature_extraction.text import CountVectorizer
from sklearn.metrics import confusion_matrix
from sklearn import metrics
from sklearn.metrics import roc_curve, auc
from nltk.stem.porter import PorterStemmer
import re
# Tutorial about Python regular expressions: https://pymotw.com/2/re/
import string
from nltk.corpus import stopwords
from nltk.stem import PorterStemmer
from nltk.stem.wordnet import WordNetLemmatizer
from gensim.models import Word2Vec
from gensim.models import KeyedVectors
import pickle
from tqdm import tqdm
import os
from plotly import plotly
import plotly.offline as offline
import plotly.graph_objs as go
offline.init_notebook_mode()
from collections import Counter
In [7]:
project_data = pd.read_csv('train_data.csv')
resource_data = pd.read_csv('resources.csv')
In [8]:
print("Number of data points in train data", project_data.shape)
print('-'*50)
print("The attributes of data :", project_data.columns.values)
In [9]:
print("Number of data points in train data", resource_data.shape)
print(resource_data.columns.values)
resource_data.head(2)
Out[9]:
In [5]:
# https://matplotlib.org/gallery/pie_and_polar_charts/pie_and_donut_labels.html#sphx-glr-gallery-pie-and-polar-charts-pie-and-donut-labels-py
y_value_counts = project_data['project_is_approved'].value_counts()
print("Number of projects thar are approved for funding ", y_value_counts[1], ", (", (y_value_counts[1]/(y_value_counts[1]+y_value_counts[0]))*100,"%)")
print("Number of projects thar are not approved for funding ", y_value_counts[0], ", (", (y_value_counts[0]/(y_value_counts[1]+y_value_counts[0]))*100,"%)")
fig, ax = plt.subplots(figsize=(6, 6), subplot_kw=dict(aspect="equal"))
recipe = ["Accepted", "Not Accepted"]
data = [y_value_counts[1], y_value_counts[0]]
wedges, texts = ax.pie(data, wedgeprops=dict(width=0.5), startangle=-40)
bbox_props = dict(boxstyle="square,pad=0.3", fc="w", ec="k", lw=0.72)
kw = dict(xycoords='data', textcoords='data', arrowprops=dict(arrowstyle="-"),
bbox=bbox_props, zorder=0, va="center")
for i, p in enumerate(wedges):
ang = (p.theta2 - p.theta1)/2. + p.theta1
y = np.sin(np.deg2rad(ang))
x = np.cos(np.deg2rad(ang))
horizontalalignment = {-1: "right", 1: "left"}[int(np.sign(x))]
connectionstyle = "angle,angleA=0,angleB={}".format(ang)
kw["arrowprops"].update({"connectionstyle": connectionstyle})
ax.annotate(recipe[i], xy=(x, y), xytext=(1.35*np.sign(x), 1.4*y),
horizontalalignment=horizontalalignment, **kw)
ax.set_title("Nmber of projects that are Accepted and not accepted")
plt.show()
In [6]:
temp = pd.DataFrame(project_data.groupby("school_state")["project_is_approved"].apply(np.mean)).reset_index()
# if you have data which contain only 0 and 1, then the mean = percentage (think about it)
temp.columns = ['state_code', 'num_proposals']
'''# How to plot US state heatmap: https://datascience.stackexchange.com/a/9620
scl = [[0.0, 'rgb(242,240,247)'],[0.2, 'rgb(218,218,235)'],[0.4, 'rgb(188,189,220)'],\
[0.6, 'rgb(158,154,200)'],[0.8, 'rgb(117,107,177)'],[1.0, 'rgb(84,39,143)']]
data = [ dict(
type='choropleth',
colorscale = scl,
autocolorscale = False,
locations = temp['state_code'],
z = temp['num_proposals'].astype(float),
locationmode = 'USA-states',
text = temp['state_code'],
marker = dict(line = dict (color = 'rgb(255,255,255)',width = 2)),
colorbar = dict(title = "% of pro")
) ]
layout = dict(
title = 'Project Proposals % of Acceptance Rate by US States',
geo = dict(
scope='usa',
projection=dict( type='albers usa' ),
showlakes = True,
lakecolor = 'rgb(255, 255, 255)',
),
)
fig = go.Figure(data=data, layout=layout)
offline.iplot(fig, filename='us-map-heat-map')
'''
Out[6]:
In [7]:
temp.sort_values(by=['num_proposals'], inplace=True)
print("States with lowest % approvals")
print(temp.head(5))
print('='*50)
print("States with highest % approvals")
print(temp.tail(5))
In [8]:
def stack_plot(data, xtick, col2='project_is_approved', col3='total'):
ind = np.arange(data.shape[0])
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, data[col3].values)
p2 = plt.bar(ind, data[col2].values)
plt.ylabel('Projects')
plt.title('Number of projects aproved vs rejected')
plt.xticks(ind, list(data[xtick].values))
plt.legend((p1[0], p2[0]), ('total', 'accepted'))
plt.show()
In [9]:
def univariate_barplots(data, col1, col2='project_is_approved', top=False):
# Count number of zeros in dataframe python: https://stackoverflow.com/a/51540521/4084039
temp = pd.DataFrame(project_data.groupby(col1)[col2].agg(lambda x: x.eq(1).sum())).reset_index()
# Pandas dataframe grouby count: https://stackoverflow.com/a/19385591/4084039
temp['total'] = pd.DataFrame(project_data.groupby(col1)[col2].agg({'total':'count'})).reset_index()['total']
temp['Avg'] = pd.DataFrame(project_data.groupby(col1)[col2].agg({'Avg':'mean'})).reset_index()['Avg']
temp.sort_values(by=['total'],inplace=True, ascending=False)
if top:
temp = temp[0:top]
stack_plot(temp, xtick=col1, col2=col2, col3='total')
print(temp.head(10))
print("="*50)
print(temp.tail(10))
In [10]:
univariate_barplots(project_data, 'school_state', 'project_is_approved', False)
In [11]:
univariate_barplots(project_data, 'teacher_prefix', 'project_is_approved' , top=False)
In [12]:
univariate_barplots(project_data, 'project_grade_category', 'project_is_approved', top=False)
In [10]:
catogories = list(project_data['project_subject_categories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())
In [11]:
project_data['clean_categories'] = cat_list
project_data.drop(['project_subject_categories'], axis=1, inplace=True)
project_data.head(2)
Out[11]:
In [15]:
univariate_barplots(project_data, 'clean_categories', 'project_is_approved', top=20)
In [16]:
from collections import Counter
my_counter = Counter()
for word in project_data['clean_categories'].values:
my_counter.update(word.split())
In [17]:
cat_dict = dict(my_counter)
sorted_cat_dict = dict(sorted(cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_cat_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_cat_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved category wise')
plt.xticks(ind, list(sorted_cat_dict.keys()))
plt.show()
In [18]:
for i, j in sorted_cat_dict.items():
print("{:20} :{:10}".format(i,j))
In [19]:
sub_catogories = list(project_data['project_subject_subcategories'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
sub_cat_list = []
for i in sub_catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp +=j.strip()+" "#" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_')
sub_cat_list.append(temp.strip())
In [20]:
project_data['clean_subcategories'] = sub_cat_list
project_data.drop(['project_subject_subcategories'], axis=1, inplace=True)
project_data.head(2)
Out[20]:
In [21]:
univariate_barplots(project_data, 'clean_subcategories', 'project_is_approved', top=50)
In [22]:
from collections import Counter
my_counter = Counter()
for word in project_data['clean_subcategories'].values:
my_counter.update(word.split())
In [23]:
sub_cat_dict = dict(my_counter)
sorted_sub_cat_dict = dict(sorted(sub_cat_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(sorted_sub_cat_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(sorted_sub_cat_dict.values()))
plt.ylabel('Projects')
plt.title('% of projects aproved state wise')
plt.xticks(ind, list(sorted_sub_cat_dict.keys()))
plt.show()
In [24]:
for i, j in sorted_sub_cat_dict.items():
print("{:20} :{:10}".format(i,j))
In [25]:
word_count = project_data['project_title'].str.split().apply(len).value_counts()
word_dict = dict(word_count)
word_dict = dict(sorted(word_dict.items(), key=lambda kv: kv[1]))
ind = np.arange(len(word_dict))
plt.figure(figsize=(20,5))
p1 = plt.bar(ind, list(word_dict.values()))
plt.ylabel('Numeber of projects')
plt.xlabel('Numeber words in project title')
plt.title('Words for each title of the project')
plt.xticks(ind, list(word_dict.keys()))
plt.show()
In [26]:
approved_title_word_count = project_data[project_data['project_is_approved']==1]['project_title'].str.split().apply(len)
approved_title_word_count = approved_title_word_count.values
rejected_title_word_count = project_data[project_data['project_is_approved']==0]['project_title'].str.split().apply(len)
rejected_title_word_count = rejected_title_word_count.values
In [27]:
plt.boxplot([approved_title_word_count, rejected_title_word_count])
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project title')
plt.grid()
plt.show()
In [28]:
plt.figure(figsize=(10,3))
sns.kdeplot(approved_title_word_count,label="Approved Projects", bw=0.6)
sns.kdeplot(rejected_title_word_count,label="Not Approved Projects", bw=0.6)
plt.legend()
plt.show()
In [12]:
project_data["essay"] = project_data["project_essay_1"].map(str) +\
project_data["project_essay_2"].map(str) + \
project_data["project_essay_3"].map(str) + \
project_data["project_essay_4"].map(str)
In [13]:
approved_word_count = project_data[project_data['project_is_approved']==1]['essay'].str.split().apply(len)
approved_word_count = approved_word_count.values
rejected_word_count = project_data[project_data['project_is_approved']==0]['essay'].str.split().apply(len)
rejected_word_count = rejected_word_count.values
In [31]:
plt.boxplot([approved_word_count, rejected_word_count])
plt.title('Words for each essay of the project')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Words in project essays')
plt.grid()
plt.show()
In [32]:
plt.figure(figsize=(10,3))
sns.distplot(approved_word_count, hist=False, label="Approved Projects")
sns.distplot(rejected_word_count, hist=False, label="Not Approved Projects")
plt.title('Words for each essay of the project')
plt.xlabel('Number of words in each eassay')
plt.legend()
plt.show()
In [33]:
resource_data.head(2)
Out[33]:
In [15]:
price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
price_data.head(2)
Out[15]:
In [16]:
project_data = pd.merge(project_data, price_data, on='id', how='left')
In [17]:
approved_price = project_data[project_data['project_is_approved']==1]['price'].values
rejected_price = project_data[project_data['project_is_approved']==0]['price'].values
In [37]:
plt.boxplot([approved_price, rejected_price])
plt.title('Box Plots of Cost per approved and not approved Projects')
plt.xticks([1,2],('Approved Projects','Rejected Projects'))
plt.ylabel('Price')
plt.grid()
plt.show()
In [40]:
plt.figure(figsize=(10,3))
sns.distplot(approved_price, hist=False, label="Approved Projects")
sns.distplot(rejected_price, hist=False, label="Not Approved Projects")
plt.title('Cost per approved and not approved Projects')
plt.xlabel('Cost of a project')
plt.legend()
plt.show()
In [18]:
from prettytable import PrettyTable
x = PrettyTable()
x.field_names = ["Percentile", "Approved Projects", "Not Approved Projects"]
for i in range(0,101,5):
x.add_row([i,np.round(np.percentile(approved_price,i), 3), np.round(np.percentile(rejected_price,i), 3)])
print(x)
In [42]:
resource_data.head(2)
Out[42]:
In [43]:
price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
price_data.head(2)
Out[43]:
1.2.9 UNIVARIATE FOR TEACHER NUMBER OF PREVIOUSLY POSTED PROJECT¶
In [44]:
univariate_barplots(project_data, 'teacher_number_of_previously_posted_projects', 'project_is_approved', top=25)
summary:¶
By this univariate anlaysis we can observe that the teacher number of previously posted projects list in top 10 and teachers
who does not propose any prior projects are also accepted.nearly 82% of the projects are approved by the teachers who did not propose any projects before, we also can see that the rate of proposal is higher if the proposed projects are atleast 20.
In [45]:
summary = []
for c in project_data["project_resource_summary"] :
summary.append(c)
summary[0:10]
Out[45]:
In [46]:
len(summary)
Out[46]:
In [47]:
numeric_summary_values = {}
for x in tqdm(range(len(summary))):
for s in summary[x].split():
if s.isdigit() :
numeric_summary_values[x] = int(s)
In [177]:
numeric_summary_values[14]
Out[177]:
In [48]:
numeric_digits = {}
for c in range(len(summary)) :
if c in numeric_summary_values.keys() :
numeric_digits[c] = numeric_summary_values[c]
else :
numeric_digits[c] = 0
In [49]:
digit_in_summary = []
for a in numeric_digits.values() :
if a > 0 :
digit_in_summary.append(1)
else :
digit_in_summary.append(0)
In [50]:
digit_in_summary[0:14]
Out[50]:
In [51]:
project_data['digit_in_summary'] = digit_in_summary
In [52]:
project_data.head(30)
Out[52]:
In [53]:
univariate_barplots(project_data, 'digit_in_summary', 'project_is_approved', top=10)
summary:¶
the summaries having numeric values contain high acceptance rate it is 90% and majority of the projects are not having numerical values stating the requirement of the products.In [54]:
project_data.head(2)
Out[54]:
In [63]:
print(project_data['project_title'].values[0])
print("="*50)
print(project_data['project_title'].values[150])
print("="*50)
print(project_data['project_title'].values[500])
print("="*50)
print(project_data['project_title'].values[1000])
print("="*50)
print(project_data['project_title'].values[2000])
print("="*50)
print(project_data['project_title'].values[5000])
print("="*50)
print(project_data['project_title'].values[7000])
print("="*50)
print(project_data['project_title'].values[12000])
print("="*50)
print(project_data['project_title'].values[17000])
print("="*50)
print(project_data['project_title'].values[50000])
print("="*50)
print(project_data['project_title'].values[98000])
print("="*50)
In [64]:
import re
def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)
# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [65]:
sent = decontracted(project_data['project_title'].values[17000])
print(sent)
print("="*50)
In [66]:
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
'won', "won't", 'wouldn', "wouldn't"]
In [67]:
preprocessed_titles = []
for titles in tqdm(project_data["project_title"]):
title = decontracted(titles)
title = title.replace('\\r', ' ')
title = title.replace('\\"', ' ')
title = title.replace('\\n', ' ')
title = re.sub('[^A-Za-z0-9]+', ' ', title)
title = ' '.join(f for f in title.split() if f not in stopwords)
preprocessed_titles.append(title.lower().strip())
In [68]:
print(preprocessed_titles[10000])
print("="*50)
print(preprocessed_titles[4444])
print("="*50)
vectorizing categorical data.¶
In [69]:
my_counter = Counter()
for state in project_data['school_state'].values:
my_counter.update(state.split())
In [70]:
school_state_cat_dict = dict(my_counter)
sorted_school_state_cat_dict = dict(sorted(school_state_cat_dict.items(), key=lambda kv: kv[1]))
In [71]:
vectorizer = CountVectorizer(vocabulary=list(sorted_school_state_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['school_state'].values)
print(vectorizer.get_feature_names())
school_state_categories_one_hot = vectorizer.transform(project_data['school_state'].values)
print("Shape of matrix after one hot encoding ",school_state_categories_one_hot.shape)
In [72]:
catogories = list(project_data['project_grade_category'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())
In [73]:
project_data['clean_grade_category'] = cat_list
project_data.drop(['project_grade_category'], axis=1, inplace=True)
project_data.head(2)
Out[73]:
In [74]:
my_counter = Counter()
for project_grade in project_data['clean_grade_category'].values:
my_counter.update(project_grade.split())
In [75]:
project_grade_cat_dict = dict(my_counter)
sorted_project_grade_cat_dict = dict(sorted(project_grade_cat_dict.items(), key=lambda kv: kv[1]))
In [76]:
vectorizer = CountVectorizer(vocabulary=list(sorted_project_grade_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_grade_category'].values)
print(vectorizer.get_feature_names())
project_grade_categories_one_hot = vectorizer.transform(project_data['clean_grade_category'].values)
print("Shape of matrix after one hot encoding ",project_grade_categories_one_hot.shape)
In [79]:
catogories = list(project_data['teacher_prefix'].values)
# remove special characters from list of strings python: https://stackoverflow.com/a/47301924/4084039
# https://www.geeksforgeeks.org/removing-stop-words-nltk-python/
# https://stackoverflow.com/questions/23669024/how-to-strip-a-specific-word-from-a-string
# https://stackoverflow.com/questions/8270092/remove-all-whitespace-in-a-string-in-python
cat_list = []
for i in catogories:
temp = ""
# consider we have text like this "Math & Science, Warmth, Care & Hunger"
for j in i.split(','): # it will split it in three parts ["Math & Science", "Warmth", "Care & Hunger"]
if 'The' in j.split(): # this will split each of the catogory based on space "Math & Science"=> "Math","&", "Science"
j=j.replace('The','') # if we have the words "The" we are going to replace it with ''(i.e removing 'The')
j = j.replace(' ','') # we are placeing all the ' '(space) with ''(empty) ex:"Math & Science"=>"Math&Science"
temp+=j.strip()+" " #" abc ".strip() will return "abc", remove the trailing spaces
temp = temp.replace('&','_') # we are replacing the & value into
cat_list.append(temp.strip())
In [80]:
project_data['clean_teacher_prefix'] = cat_list
project_data.drop(['teacher_prefix'], axis=1, inplace=True)
project_data.head(2)
Out[80]:
In [78]:
project_data["teacher_prefix"].fillna(" ", inplace = True)
In [81]:
my_counter = Counter()
for teacher_prefix in project_data['clean_teacher_prefix'].values:
teacher_prefix = str(teacher_prefix)
my_counter.update(teacher_prefix.split())
In [82]:
teacher_prefix_cat_dict = dict(my_counter)
sorted_teacher_prefix_cat_dict = dict(sorted(teacher_prefix_cat_dict.items(), key=lambda kv: kv[1]))
In [83]:
#https://stackoverflow.com/questions/39303912/tfidfvectorizer-in-scikit-learn-valueerror-np-nan-is-an-invalid-document/39308809#39308809
vectorizer = CountVectorizer(vocabulary=list(sorted_teacher_prefix_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_teacher_prefix'].values.astype("U"))
#teacher_prefix_new = project_data['teacher_prefix'].notnull()
print(vectorizer.get_feature_names())
teacher_prefix_categories_one_hot = vectorizer.transform(project_data['clean_teacher_prefix'].values.astype("U"))
print("Shape of matrix after one hot encoding ",teacher_prefix_categories_one_hot.shape)
1.4.2.2 BOW ON PROJECT TITLE¶
In [84]:
# merge two column text dataframe:
project_data["essay"] = project_data["project_essay_1"].map(str) +\
project_data["project_essay_2"].map(str) + \
project_data["project_essay_3"].map(str) + \
project_data["project_essay_4"].map(str)
In [85]:
# https://stackoverflow.com/a/47091490/4084039
import re
def decontracted(phrase):
# specific
phrase = re.sub(r"won't", "will not", phrase)
phrase = re.sub(r"can\'t", "can not", phrase)
# general
phrase = re.sub(r"n\'t", " not", phrase)
phrase = re.sub(r"\'re", " are", phrase)
phrase = re.sub(r"\'s", " is", phrase)
phrase = re.sub(r"\'d", " would", phrase)
phrase = re.sub(r"\'ll", " will", phrase)
phrase = re.sub(r"\'t", " not", phrase)
phrase = re.sub(r"\'ve", " have", phrase)
phrase = re.sub(r"\'m", " am", phrase)
return phrase
In [86]:
sent = decontracted(project_data['essay'].values[20000])
print(sent)
print("="*50)
In [87]:
# \r \n \t remove from string python: http://texthandler.com/info/remove-line-breaks-python/
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
print(sent)
In [88]:
# https://gist.github.com/sebleier/554280
# we are removing the words from the stop words list: 'no', 'nor', 'not'
stopwords= ['i', 'me', 'my', 'myself', 'we', 'our', 'ours', 'ourselves', 'you', "you're", "you've",\
"you'll", "you'd", 'your', 'yours', 'yourself', 'yourselves', 'he', 'him', 'his', 'himself', \
'she', "she's", 'her', 'hers', 'herself', 'it', "it's", 'its', 'itself', 'they', 'them', 'their',\
'theirs', 'themselves', 'what', 'which', 'who', 'whom', 'this', 'that', "that'll", 'these', 'those', \
'am', 'is', 'are', 'was', 'were', 'be', 'been', 'being', 'have', 'has', 'had', 'having', 'do', 'does', \
'did', 'doing', 'a', 'an', 'the', 'and', 'but', 'if', 'or', 'because', 'as', 'until', 'while', 'of', \
'at', 'by', 'for', 'with', 'about', 'against', 'between', 'into', 'through', 'during', 'before', 'after',\
'above', 'below', 'to', 'from', 'up', 'down', 'in', 'out', 'on', 'off', 'over', 'under', 'again', 'further',\
'then', 'once', 'here', 'there', 'when', 'where', 'why', 'how', 'all', 'any', 'both', 'each', 'few', 'more',\
'most', 'other', 'some', 'such', 'only', 'own', 'same', 'so', 'than', 'too', 'very', \
's', 't', 'can', 'will', 'just', 'don', "don't", 'should', "should've", 'now', 'd', 'll', 'm', 'o', 're', \
've', 'y', 'ain', 'aren', "aren't", 'couldn', "couldn't", 'didn', "didn't", 'doesn', "doesn't", 'hadn',\
"hadn't", 'hasn', "hasn't", 'haven', "haven't", 'isn', "isn't", 'ma', 'mightn', "mightn't", 'mustn',\
"mustn't", 'needn', "needn't", 'shan', "shan't", 'shouldn', "shouldn't", 'wasn', "wasn't", 'weren', "weren't", \
'won', "won't", 'wouldn', "wouldn't"]
In [89]:
from tqdm import tqdm
preprocessed_essays = []
# tqdm is for printing the status bar
for sentance in tqdm(project_data['essay'].values):
sent = decontracted(sentance)
sent = sent.replace('\\r', ' ')
sent = sent.replace('\\"', ' ')
sent = sent.replace('\\n', ' ')
sent = re.sub('[^A-Za-z0-9]+', ' ', sent)
# https://gist.github.com/sebleier/554280
sent = ' '.join(e for e in sent.split() if e not in stopwords)
preprocessed_essays.append(sent.lower().strip())
In [91]:
vectorizer = CountVectorizer(min_df=10)
text_bow = vectorizer.fit_transform(preprocessed_essays)
print("Shape of matrix after one hot encoding ",text_bow.shape)
In [92]:
preprocessed_titles = []
for titles in tqdm(project_data["project_title"]):
title = decontracted(titles)
title = title.replace('\\r', ' ')
title = title.replace('\\"', ' ')
title = title.replace('\\n', ' ')
title = re.sub('[^A-Za-z0-9]+', ' ', title)
title = ' '.join(f for f in title.split() if f not in stopwords)
preprocessed_titles.append(title.lower().strip())
In [94]:
vectorizer = CountVectorizer(min_df=10)
text_bow = vectorizer.fit_transform(preprocessed_titles)
print("Shape of matrix after one hot encoding ",text_bow.shape)
'''TFIDF on Project_title:'''¶
In [95]:
from sklearn.feature_extraction.text import TfidfVectorizer
vectorizer = TfidfVectorizer(min_df=10)
text_tfidf = vectorizer.fit_transform(preprocessed_essays)
print("Shape of matrix after one hot encodig ",text_tfidf.shape)
In [96]:
from sklearn.feature_extraction.text import TfidfVectorizer
vectorizer = TfidfVectorizer(min_df=20)
text_tfidf = vectorizer.fit_transform(preprocessed_titles)
print("Shape of matrix after one hot encodig ",text_tfidf.shape)
In [101]:
'''
# Reading glove vectors in python: https://stackoverflow.com/a/38230349/4084039
def loadGloveModel(gloveFile):
print ("Loading Glove Model")
f = open(gloveFile,'r', encoding="utf8")
model = {}
for line in tqdm(f):
splitLine = line.split()
word = splitLine[0]
embedding = np.array([float(val) for val in splitLine[1:]])
model[word] = embedding
print ("Done.",len(model)," words loaded!")
return model
model = loadGloveModel('glove.42B.300d.txt')
# ============================
Output:
Loading Glove Model
1917495it [06:32, 4879.69it/s]
Done. 1917495 words loaded!
# ============================
words = []
for i in preproced_texts:
words.extend(i.split(' '))
for i in preproced_titles:
words.extend(i.split(' '))
print("all the words in the coupus", len(words))
words = set(words)
print("the unique words in the coupus", len(words))
inter_words = set(model.keys()).intersection(words)
print("The number of words that are present in both glove vectors and our coupus", \
len(inter_words),"(",np.round(len(inter_words)/len(words)*100,3),"%)")
words_courpus = {}
words_glove = set(model.keys())
for i in words:
if i in words_glove:
words_courpus[i] = model[i]
print("word 2 vec length", len(words_courpus))
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
import pickle
with open('glove_vectors', 'wb') as f:
pickle.dump(words_courpus, f)
'''
Out[101]:
In [165]:
# stronging variables into pickle files python: http://www.jessicayung.com/how-to-use-pickle-to-save-and-load-variables-in-python/
# make sure you have the glove_vectors file
with open('glove_vectors', 'rb') as f:
model = pickle.load(f)
glove_words = set(model.keys())
In [166]:
avg_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_vectors.append(vector)
print(len(avg_w2v_vectors))
print(len(avg_w2v_vectors[0]))
'''avg w2v on project_title'''¶
In [167]:
avg_w2v_vectors_titles = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
cnt_words =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if word in glove_words:
vector += model[word]
cnt_words += 1
if cnt_words != 0:
vector /= cnt_words
avg_w2v_vectors_titles.append(vector)
print(len(avg_w2v_vectors_titles))
print(len(avg_w2v_vectors_titles[0]))
In [168]:
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_essays)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [107]:
tfidf_w2v_vectors = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_essays): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_vectors.append(vector)
print(len(tfidf_w2v_vectors))
print(len(tfidf_w2v_vectors[0]))
'''tfidf weighted w2v on project_title'''¶
In [169]:
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_titles)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [170]:
tfidf_w2v_vectors_titles = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_vectors_titles.append(vector)
print(len(tfidf_w2v_vectors_titles))
print(len(tfidf_w2v_vectors_titles[0]))
In [115]:
# we get the cost of the project using resource.csv file
resource_data.head(2)
Out[115]:
In [121]:
# https://stackoverflow.com/questions/22407798/how-to-reset-a-dataframes-indexes-for-all-groups-in-one-step
price_data = resource_data.groupby('id').agg({'price':'sum', 'quantity':'sum'}).reset_index()
price_data.head(2)
Out[121]:
In [122]:
# join two dataframes in python:
project_data = pd.merge(project_data, price_data, on='id', how='left')
In [123]:
approved_price = project_data[project_data['project_is_approved']==1]['price'].values
rejected_price = project_data[project_data['project_is_approved']==0]['price'].values
In [124]:
# check this one: https://www.youtube.com/watch?v=0HOqOcln3Z4&t=530s
# standardization sklearn: https://scikit-learn.org/stable/modules/generated/sklearn.preprocessing.StandardScaler.html
from sklearn.preprocessing import StandardScaler
price_scalar = StandardScaler()
price_scalar.fit(project_data['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print(f"Mean : {price_scalar.mean_[0]}, Standard deviation : {np.sqrt(price_scalar.var_[0])}")
# Now standardize the data with above maen and variance.
price_standardized = price_scalar.transform(project_data['price'].values.reshape(-1, 1))
price_standardized
Out[124]:
In [126]:
from sklearn.feature_extraction.text import CountVectorizer
vectorizer = CountVectorizer(vocabulary=list(sorted_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_categories'].values)
print(vectorizer.get_feature_names())
categories_one_hot = vectorizer.transform(project_data['clean_categories'].values)
print("Shape of matrix after one hot encoding ",categories_one_hot.shape)
In [127]:
'''vectorizing categorical data-clean sub categories of projects'''
vectorizer = CountVectorizer(vocabulary=list(sorted_sub_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_subcategories'].values)
print(vectorizer.get_feature_names())
sub_categories_one_hot = vectorizer.transform(project_data['clean_subcategories'].values)
print("Shape of matrix after one hot encoding ",sub_categories_one_hot.shape)
'''one hot code for --school states'''¶
In [128]:
my_counter = Counter()
for state in project_data['school_state'].values:
my_counter.update(state.split())
school_state_cat_dict = dict(my_counter)
sorted_school_state_cat_dict = dict(sorted(school_state_cat_dict.items(), key=lambda kv: kv[1]))
In [129]:
vectorizer = CountVectorizer(vocabulary=list(sorted_school_state_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['school_state'].values)
print(vectorizer.get_feature_names())
school_state_categories_one_hot = vectorizer.transform(project_data['school_state'].values)
print("Shape of matrix after one hot encoding ",school_state_categories_one_hot.shape)
'''one hot code for project_grade_category'''¶
In [130]:
my_counter = Counter()
for project_grade in project_data['clean_grade_category'].values:
my_counter.update(project_grade.split())
project_grade_cat_dict = dict(my_counter)
sorted_project_grade_cat_dict = dict(sorted(project_grade_cat_dict.items(), key=lambda kv: kv[1]))
In [141]:
vectorizer = CountVectorizer(vocabulary=list(sorted_project_grade_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_grade_category'].values)
print(vectorizer.get_feature_names())
clean_grade_categories_one_hot = vectorizer.transform(project_data['clean_grade_category'].values)
print("Shape of matrix after one hot encoding ",clean_grade_categories_one_hot.shape)
'''one hot code for --teacher prefix'''¶
In [143]:
my_counter = Counter()
for teacher_prefix in project_data['clean_teacher_prefix'].values:
teacher_prefix = str(teacher_prefix)
my_counter.update(teacher_prefix.split())
teacher_prefix_cat_dict = dict(my_counter)
sorted_teacher_prefix_cat_dict = dict(sorted(teacher_prefix_cat_dict.items(), key=lambda kv: kv[1]))
In [144]:
vectorizer = CountVectorizer(vocabulary=list(sorted_teacher_prefix_cat_dict.keys()), lowercase=False, binary=True)
vectorizer.fit(project_data['clean_teacher_prefix'].values.astype("U"))
print(vectorizer.get_feature_names())
clean_teacher_prefix_categories_one_hot = vectorizer.transform(project_data['clean_teacher_prefix'].values.astype("U"))
print("Shape of matrix after one hot encoding ",clean_teacher_prefix_categories_one_hot.shape)
In [134]:
print(categories_one_hot.shape)
print(sub_categories_one_hot.shape)
print(text_bow.shape)
print(price_standardized.shape)
In [135]:
# merge two sparse matrices: https://stackoverflow.com/a/19710648/4084039
from scipy.sparse import hstack
# with the same hstack function we are concatinating a sparse matrix and a dense matirx :)
X = hstack((categories_one_hot, sub_categories_one_hot, text_bow, price_standardized))
X.shape
Out[135]:
In [136]:
from sklearn.preprocessing import StandardScaler
price_scalar = StandardScaler()
price_scalar.fit(project_data['price'].values.reshape(-1,1)) # finding the mean and standard deviation of this data
print("Mean : {}".format(price_scalar.mean_[0]))
print("Standard deviation : {}".format(np.sqrt(price_scalar.var_[0])))
# Now standardize the data with above maen and variance.
price_standardized = price_scalar.transform(project_data['price'].values.reshape(-1, 1))
In [137]:
import warnings
warnings.filterwarnings("ignore")
prev_projects_scalar = StandardScaler()
prev_projects_scalar.fit(project_data['teacher_number_of_previously_posted_projects'].values.reshape(-1,1))
print("Mean : {}".format(prev_projects_scalar.mean_[0]))
print("Standard deviation : {}".format(np.sqrt(prev_projects_scalar.var_[0])))
prev_projects_standardized = prev_projects_scalar.transform(project_data['teacher_number_of_previously_posted_projects'].
values.reshape(-1, 1))
prev_projects_standardized
Out[137]:
In [138]:
vectorizer = CountVectorizer(min_df=5)
title_bow = vectorizer.fit_transform(preprocessed_titles)
print("Shape of matrix after one hot encoding ",title_bow.shape)
In [139]:
import warnings
warnings.filterwarnings("ignore")
quantity_scalar = StandardScaler()
## Finding the mean and standard deviation of this data
quantity_scalar.fit(project_data['quantity'].values.reshape(-1,1))
print("Mean : {}".format(quantity_scalar.mean_[0]))
print("Standard deviation : {}".format(np.sqrt(quantity_scalar.var_[0])))
# Now standardize the data with above maen and variance.
quantity_standardized = quantity_scalar.transform(project_data['quantity'].values.reshape(-1, 1))
'''tsne Bag of words encoding project title feature'''¶
In [145]:
X = hstack((categories_one_hot, sub_categories_one_hot,
school_state_categories_one_hot,
clean_grade_categories_one_hot,
clean_teacher_prefix_categories_one_hot,
price_standardized,
quantity_standardized,
prev_projects_standardized,
title_bow))
X.shape
Out[145]:
In [146]:
from sklearn.manifold import TSNE
X = X.tocsr()
X_new = X[0:5000,:]
X_new = X_new.toarray()
model = TSNE(n_components = 2, perplexity = 100.0, random_state = 0)
tsne_data_b = model.fit_transform(X_new)
labels = project_data["project_is_approved"]
labels_new = labels[0: 5000]
len(labels_new)
Out[146]:
we can see that the data is completely overlapped, and the points are scattered and formed into group of clusters.¶
In [147]:
tsne_data_b = np.vstack((tsne_data_b.T, labels_new)).T
tsne_df_b = pd.DataFrame(tsne_data_b, columns = ("1st_Dim","2nd_Dim","Labels"))
tsne_df_b.shape
sns.FacetGrid(tsne_df_b, hue = "Labels", size = 10).map(plt.scatter, "1st_Dim", "2nd_Dim").add_legend().fig.suptitle("TSNE WITH BOW ENCODING OF PROJECT TITLE FEATURE ")
plt.show()
'''tsne for tfidf encoding project title'''¶
In [149]:
'''tsne for tfidf encoding project title'''
X = hstack((categories_one_hot,
sub_categories_one_hot,
school_state_categories_one_hot,
clean_grade_categories_one_hot,
clean_teacher_prefix_categories_one_hot,
price_standardized,
quantity_standardized,
prev_projects_standardized,
text_tfidf))
X.shape
Out[149]:
In [151]:
X = X.tocsr()
X_new = X[0:5000,:]
X_new = X_new.toarray()
model = TSNE(n_components = 2, perplexity = 100.0, random_state = 0)
tsne_data_tfidf = model.fit_transform(X_new)
tsne_data_tfidf = np.vstack((tsne_data_tfidf.T, labels_new)).T
tsne_df_tfidf = pd.DataFrame(tsne_data_tfidf, columns = ("1st_Dim","2nd_Dim","Labels"))
tsne_df_tfidf.shape
sns.FacetGrid(tsne_df_tfidf, hue = "Labels", size = 10).map(plt.scatter, "1st_Dim", "2nd_Dim").add_legend().fig.suptitle("TSNE WITH TF-IDF ENCODING OF PROJECT TITLE FEATURE ")
plt.show()
we can observe that the points are scattered in different directions and cannot seem to find the conclusion¶
tsne for avg w2v encoding project title¶
In [152]:
'''tsne for avg w2v encoding project title'''
X = hstack((categories_one_hot,
sub_categories_one_hot,
school_state_categories_one_hot,
project_grade_categories_one_hot,
teacher_prefix_categories_one_hot,
price_standardized,
quantity_standardized,
prev_projects_standardized,
avg_w2v_vectors_titles))
X.shape
Out[152]:
In [153]:
X = X.tocsr()
X_new = X[0:5000,:]
X_new = X_new.toarray()
model = TSNE(n_components = 2, perplexity = 100.0, random_state = 0)
tsne_data_avg_w2v = model.fit_transform(X_new)
In [154]:
tsne_data_avg_w2v = np.vstack((tsne_data_avg_w2v.T, labels_new)).T
tsne_df_avg_w2v = pd.DataFrame(tsne_data_avg_w2v, columns = ("1st_Dim","2nd_Dim","Labels"))
In [155]:
tsne_df_avg_w2v.shape
Out[155]:
In [156]:
sns.FacetGrid(tsne_df_avg_w2v, hue = "Labels", size = 10).map(plt.scatter, "1st_Dim", "2nd_Dim").add_legend().fig.suptitle("TSNE WITH AVG W2V ENCODING OF PROJECT TITLE FEATURE ")
plt.show()
In [172]:
#https://stackoverflow.com/questions/27431390/typeerror-zip-object-is-not-subscriptable
def text_from_tagged_ngram(ngram):
if type(ngram) == tuple:
return ngram[0]
return " ".join(zip(*ngram)[0])
this tsne of average w2v and project title we cannot observe any clusters and also cannot find weather the project is accepted or not¶
In [173]:
tfidf_model = TfidfVectorizer()
tfidf_model.fit(preprocessed_titles)
# we are converting a dictionary with word as a key, and the idf as a value
dictionary = dict(zip(tfidf_model.get_feature_names(), list(tfidf_model.idf_)))
tfidf_words = set(tfidf_model.get_feature_names())
In [174]:
tfidf_w2v_vectors_title = []; # the avg-w2v for each sentence/review is stored in this list
for sentence in tqdm(preprocessed_titles): # for each review/sentence
vector = np.zeros(300) # as word vectors are of zero length
tf_idf_weight =0; # num of words with a valid vector in the sentence/review
for word in sentence.split(): # for each word in a review/sentence
if (word in glove_words) and (word in tfidf_words):
vec = model[word] # getting the vector for each word
# here we are multiplying idf value(dictionary[word]) and the tf value((sentence.count(word)/len(sentence.split())))
tf_idf = dictionary[word]*(sentence.count(word)/len(sentence.split())) # getting the tfidf value for each word
vector += (vec * tf_idf) # calculating tfidf weighted w2v
tf_idf_weight += tf_idf
if tf_idf_weight != 0:
vector /= tf_idf_weight
tfidf_w2v_vectors_title.append(vector)
print(len(tfidf_w2v_vectors_title))
print(len(tfidf_w2v_vectors_title[0]))
'''TSNE with TFIDF Weighted W2V encoding of project_title feature'''¶
In [175]:
X = hstack((categories_one_hot,
sub_categories_one_hot,
school_state_categories_one_hot,
clean_grade_categories_one_hot,
clean_teacher_prefix_categories_one_hot,
price_standardized,
quantity_standardized,
prev_projects_standardized,
tfidf_w2v_vectors_title))
X.shape
Out[175]:
In [176]:
X = X.tocsr()
X_new = X[0:5000,:]
X_new = X_new.toarray()
model = TSNE(n_components = 2, perplexity = 100.0, random_state = 0)
tsne_data_tfidf_w2v = model.fit_transform(X_new)
tsne_data_tfidf_w2v = np.vstack((tsne_data_tfidf_w2v.T, labels_new)).T
tsne_df_tfidf_w2v = pd.DataFrame(tsne_data_tfidf_w2v, columns = ("1st_Dim","2nd_Dim","Labels"))
tsne_df_tfidf_w2v.shape
sns.FacetGrid(tsne_df_tfidf_w2v, hue = "Labels", size = 10).map(plt.scatter, "1st_Dim", "2nd_Dim").add_legend().fig.suptitle("TSNE WITH TF-IDF WEIGHTED W2V ENCODING OF PROJECT TITLE FEATURE ")
plt.show()
these combinations are not clearly drawing us through the results and we might have to try more combinations for desired result.¶
In [178]:
X = hstack((categories_one_hot, sub_categories_one_hot, school_state_categories_one_hot,
project_grade_categories_one_hot, teacher_prefix_categories_one_hot, price_standardized,
quantity_standardized, prev_projects_standardized, title_bow, text_tfidf, avg_w2v_vectors_titles,
tfidf_w2v_vectors_title))
X.shape
Out[178]:
In [190]:
X = X.tocsr()
X_new = X[0:5000,:]
tsne_data_complete = model.fit_transform(X_new.toarray())
In [191]:
tsne_data_complete= np.vstack((tsne_data_complete.T, labels_new)).T
tsne_df_complete = pd.DataFrame(tsne_data_complete, columns = ("1st_Dim","2nd_Dim","Labels"))
In [192]:
tsne_df_complete.shape
Out[192]:
In [198]:
sns.FacetGrid(tsne_df_complete, hue = "Labels", size = 10).map(plt.scatter, "1st_Dim", "2nd_Dim").add_legend().fig.suptitle("TSNE WITH BOW, TF-IDF, AVG W2V, TF-IDF WEIGHTED W2V ENCODING OF PROJECT TITLE FEATURE ")
plt.show()
These combinations are not clearly drawing us through the results and we might have to try more combinations for desired result of clustering similar data points.
conclusion:
1. DE state from united states is having highest percentage of projects which are accepted in the whole country and almost it has 90% of the acceptance rate. and ND is having 89% and WA is having 88% respectively.
2.VT is having the lowest approval rate with 80% followed by DC and TX with 80 % and 81%.
3.female teachers are having the max number of projects proposed and accepted when compared to the male teachers.
4.there are lot of projects proposed for pre kindergarden adn 2nd grade while the rest it is decreasing as the increase in grade can be seen.
5.We can also notice that the Students between the 9th Grade and 12th Grade have the lowest number of projects proposed and accepted.
6. Projects belonging to Literacy and Language categories have the highest number of projects proposed and The maximum number of projects that are accepted also belong to the same category, with the acceptance rate of 87%.
7. Projects belonging to Maths and Science are having acceptance rate of 82% while the concept of Literacy and Language to this can increase its accpetance rate to 87%
8. Projects belonging to both Maths and Science when combined with the Applied Learning is having the least number of projects proposed and approved.
9. There is Variability in Acceptance rate, projects under the category Warmth, Care and Hunger have an accpetance rate of 93.5%
10. The highest number of projects are registered under Literacy and Langauage with 52,239 projects, followed by Maths and Science having 41,421 projects.
11. The sub-Category Literacy has the highest number of projects approved with 8371 projects. Also the accpetance rate is 88%.
12. The sub-Category Health and Wellness have the lowest number of projects proposed with 3,583 projects only.
13. Roughly most of the projects have 3, 4 or 5 words in the title.There are hardly any project titles containing more than 10 words.
14. The number of words in the Project Essays of Approved Projects are slightly more than the number of words in the Project Essays of the Rejected Projects.
15. The Maximum price for any project should be less than 10,000 dollars.The approved projects tend to have lower cost when compared to the projects that have not been approved.
16. We observe that it is not mandatory for a teacher to have proposed any project prior. Maximum number of teachers, nearly 82% of the approved projects have been submitted by teachers with no prior project proposals. New talent and efforts are well appreciated.
17. Very few teachers who have proposed more than 20 projects have got approval. But the rate of approval is Higher given the teacher has proposed atleast 19 different projects.
18. The project summaries containing numeric values have a very high acceptance rate of 90%. Well, proper numbered requirements suggest clarity in the proposals and hence Alot of people tend to donate for a better cause, that is to help children.
19. We observe that on an average Each project costs nearly 298 Dollars. The Price paid is generally for the purchase of the Items. The projects on an average require atleast 17 Different of similar items.
20. Visualisation of TSNE with Bag of Words, TF-IDF, Avg Word2Vec, TF-IDF Weighted Word2Vec does not seem to yield the expected result of clustering similar data points. Hence we would have to try any other method.
This is all about the donars choose data set and it analysis. the above drawn conclusions are drawn from the observations done in the sample data set and also by performing some of the operations on the given data.